WO2011149983A2 - Method and apparatus for trusted execution in infrastructure as a service cloud environments - Google Patents

Method and apparatus for trusted execution in infrastructure as a service cloud environments

Info

Publication number
WO2011149983A2
WO2011149983A2 PCT/US2011/037799 US2011037799W WO2011149983A2 WO 2011149983 A2 WO2011149983 A2 WO 2011149983A2 US 2011037799 W US2011037799 W US 2011037799W WO 2011149983 A2 WO2011149983 A2 WO 2011149983A2
Authority
WO
Grant status
Application
Patent type
Prior art keywords
session key
processing core
cache
encrypt
vm
Prior art date
Application number
PCT/US2011/037799
Other languages
French (fr)
Other versions
WO2011149983A3 (en )
Inventor
Fabio R. Maino
Pere Monclus
Original Assignee
Cisco Technology, Inc.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L9/00Cryptographic mechanisms or cryptographic arrangements for secret or secure communication
    • H04L9/08Key distribution or management, e.g. generation, sharing or updating, of cryptographic keys or passwords
    • H04L9/0894Escrow, recovery or storing of secret information, e.g. secret key escrow or cryptographic key storage
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/50Monitoring users, programs or devices to maintain the integrity of platforms, e.g. of processors, firmware or operating systems
    • G06F21/52Monitoring users, programs or devices to maintain the integrity of platforms, e.g. of processors, firmware or operating systems during program execution, e.g. stack integrity ; Preventing unwanted data erasure; Buffer overflow
    • G06F21/53Monitoring users, programs or devices to maintain the integrity of platforms, e.g. of processors, firmware or operating systems during program execution, e.g. stack integrity ; Preventing unwanted data erasure; Buffer overflow by executing in a restricted environment, e.g. sandbox or secure virtual machine
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/70Protecting specific internal or peripheral components, in which the protection of a component leads to protection of the entire computer
    • G06F21/71Protecting specific internal or peripheral components, in which the protection of a component leads to protection of the entire computer to assure secure computing or processing of information
    • G06F21/72Protecting specific internal or peripheral components, in which the protection of a component leads to protection of the entire computer to assure secure computing or processing of information in cryptographic circuits
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/70Protecting specific internal or peripheral components, in which the protection of a component leads to protection of the entire computer
    • G06F21/78Protecting specific internal or peripheral components, in which the protection of a component leads to protection of the entire computer to assure secure storage of data
    • G06F21/79Protecting specific internal or peripheral components, in which the protection of a component leads to protection of the entire computer to assure secure storage of data in semiconductor storage media, e.g. directly-addressable memories
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44Arrangements for executing specific programs
    • G06F9/455Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
    • G06F9/45533Hypervisors; Virtual machine monitors
    • G06F9/45558Hypervisor-specific management and integration aspects
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44Arrangements for executing specific programs
    • G06F9/455Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
    • G06F9/45533Hypervisors; Virtual machine monitors
    • G06F9/45558Hypervisor-specific management and integration aspects
    • G06F2009/45587Isolation or security of virtual machine instances

Abstract

The present disclosure presents a method and apparatus configured to provide for the trusted execution of virtual machines (VMs) on a virtualization server, e.g., for executing VMs on a virtualization server provided within Infrastructure as a Service (IaaS) cloud environment. A physical multi-core CPU may be configured with a hardware trust anchor. The trust anchor itself may be configured to manage session keys used to encrypt/decrypt instructions and data when a VM (or hypervisor) is executed on one of the CPU cores. When a context switch occurs due to an exception, the trust anchor swaps the session key used to encrypt/decrypt the contents of memory and cache allocated to a VM (or hypervisor).

Description

METHOD AND APPARATUS FOR TRUSTED EXECUTION IN

INFRASTRUCTURE AS A SERVICE CLOUD ENVIRONMENTS

TECHNICAL FIELD

[0001] Embodiments presented in this disclosure generally relate to techniques for providing a computing infrastructure for a virtualization server, and more particularly, to a method and apparatus for trusted virtual machine execution in an Infrastructure as a Service (laaS) cloud environment.

BACKGROUND

[0002] Server virtualization technology allows multiple virtual machines to run concurrently on a single physical computing system. Currently, data center environments are used to create large clusters of such physical computing systems (commonly referred to as servers), where each server runs multiple virtual machines (VMs). This approach has led to data centers that can supply massive amounts of computing power. Several providers currently allow users to supply virtual machine instances to run on the virtualization servers provided by the operator of the data center. In various forms, this general model of computing has come to be referred to as "cloud computing" or "Infrastructure as a Service" (laaS) because users simply run their virtual machine instances on an abstract hardware platform, without having to own or manage that hardware platform. This approach allows a given user to rapidly scale up dozens, if not hundreds or thousands of virtual machine instances to respond to changes in demand for computing resources.

[0003] At the same time, a significant obstacle faced by of cloud computing and laaS providers is the need for users to trust the cloud provider and the implementation of the hypervisor (or Virtual Machine Manager, (VMM)) that manages multiple VMs running on a server. Currently this is done by audits and understanding of the operating processes that cloud providers have in place in regards to security. This leads to a sense of trust, but there is not a strong mechanism to guarantee effective separation between VMs executed in the same central processing unit (CPU) server socket.

[0004] The fact that VMs belonging to different enterprises are collocated in the same server can lead to side attacks and security concerns. For example, when VMs belonging to different enterprises are executed in distinct cores of the same CPU,

1830343J 1 they share access to memory via an L3 cache (or bypassing memory protections set by the hypervisor). In the hands of skilled individuals, this could lead to information being leaked from one VM to the other. That is, a malicious VM from one enterprise could try to gain access the memory of a VM running on behalf of another enterprise.

[0005] Another concern is the hypervisor itself. Even if the hypervisor is designed to enforce separation between distinct VMs, data could still leak from one VM to the other, either for a hypervisor bug or for explicit attacks directed against the hypervisor itself. Further, another set of attacks could be launched by a malicious administrator who is part of the cloud provider personnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] So that the manner in which the above-recited features of the present disclosure can be understood in detail, a more particular description of the disclosure, briefly summarized above, may be had by reference to embodiments, some of which are illustrated in the appended drawings. It is to be noted, however, that the appended drawings illustrate only typical embodiments of this disclosure and are therefore not to be considered limiting of its scope, for the disclosure may admit to other equally effective embodiments.

[0007] Figures 1A-1 B are block diagrams illustrating different security models for trusted execution in Infrastructure as a Service (laaS) cloud environments, according to certain embodiments of the present disclosure.

[0008] Figure 2 is a block diagram illustrating components of a virtualization server configured to provide for the trusted execution of multiple VMs in an laaS environment, according to certain embodiments of the present disclosure.

[0009] Figure 3 is a block diagram illustrating an example of a processor architecture configured to provide for trusted virtual machine execution in an laaS environment, according to certain embodiments of the present disclosure.

[0010] Figure 4 further illustrates aspects of the CPU architecture first shown in Figure 3, according to certain embodiments of the present disclosure.

[0011] Figure 5 illustrates a method for securely booting a VM instance, according to certain embodiments of the present disclosure.

1830343J 2 [0012] Figure 6 illustrates a method for performing a context switch on a processing core of a multi-core CPU executing multiple VM instances, according to certain embodiments of the present disclosure.

DESCRIPTION OVERVIEW

[0013] One embodiment described herein includes a method for performing a context switch on a processing core of a CPU. The method may generally include receiving an indication of an incoming execution context for the processing core and storing a first session key. The first session key is associated with a current execution context of the processing core. The method may also include retrieving a second session key, where the second session key is associated with the incoming execution context. The method may also include initializing an encrypt/decrypt block on the CPU using the second session key. The encrypt/decrypt block is disposed between the processing core and a cache accessed by the processing core. [0014] In a particular embodiment, the method may further include the incoming execution context on the processing core. In such a case, the encrypt/decrypt block decrypts data read from the cache and encrypts data written to the cache using the second session key. The encrypt/decrypt block itself may provide a hardware implementation of a stream cipher, and wherein a key stream is generated for a read from the cache using the second session key, a nonce value, and a memory address associated with a read or write operation.

[0015] Another embodiment of the disclosure includes a system having a memory and a central processing unit (CPU). The CPU may generally include a processing core, a cache and a trust anchor. The trust anchor may be generally configured to receive an indication of an incoming execution context for the processing core, store a first session key, where the first session key is associated with a current execution context of the processing core. The trust anchor may be further configured to retrieve a second session key, wherein the second session key is associated with the incoming execution context and initialize an encrypt/decrypt block on the CPU using the second session key, wherein the encrypt/decrypt block is disposed between the processing core and a cache accessed by the processing core.

1830343J 3 [0016] Still another embodiment includes a method for executing a plurality of virtual machine instances on a multi-core CPU. This method may generally include determining an exception has occurred while executing a first virtual machine instance on a first processing core of a multi-core CPU, signaling a trust anchor coupled to the CPU with an indication of an incoming execution context for the first processing core, and storing a first session key. The first session key is associated the first virtual machine executed on the processing core. This method may also include retrieving a second session key, where the second session key is associated with the incoming execution context and initializing an encrypt/decrypt block on the CPU using the second session key, wherein the encrypt/decrypt block is disposed between the first processing core and a cache shared by each processing core on the multi-core CPU.

DESCRIPTION OF EXAMPLE EMBODIMENTS

[0017] Embodiments described herein present a method and apparatus configured to provide for the trusted execution of virtual machines VMs on a virtualization server, e.g., for executing VMs on a virtualization server provided within Infrastructure as a Service (laaS) cloud environment. One embodiment described herein provides a model for memory compartmentalization between VM instances executed on the VM server. The model may allow VMs to be executed in an un-trusted cloud provider infrastructure in a trusted manner. In particular, the memory compartmentalization techniques described herein allow a virtualization server to boot an encrypted/signed VM image as well as prevent other entities running on that virtualization server from accessing a given VM's memory, even in cases where an entity is executing running with higher execution privileges {e.g., the hypervisor). [0018] Further, embodiments described herein provide an infrastructure that may be used to provision the appropriate credentials (i.e., an encryption key) to enable trusted execution of VMs on a CPU core. Doing so may protect the memory from being leaked (maliciously or otherwise) to VMs running on other cores in cases where the VMs share accesses to a common L3 cache.

[0019] As described in greater detail herein, a physical multi-core CPU may include a hardware trust anchor component. The trust anchor may be configured to manage keys used to encrypt/decrypt instructions and data when a VM or hypervisor

1830343J 4 is executed on a core of the multi-core CPU. In one embodiment, the trust anchor may receive a key associated with an encrypted VM boot image using a secure communications process. Once received and stored by the trust anchor, the key may be used to decrypt the VM boot image and load it into memory on the virtualization server. When the hypervisor performs a context switch to execute the resulting VM instance, the trust anchor provides a session key (along with other data) used to encrypt/decrypt memory read and write operations.

[0020] In one embodiment, an encrypt/decrypt block on each cache line between a shared L3 cache and an L2 cache local to each processing core may operate in an Advanced Encryption Standard (AES) Counter Mode (AES-CTR) using the session key supplied by the trust anchor, a memory address, and a nonce to encrypt data written into the L3 cache. The same information {e.g., the key, nonce and memory address) is used to decrypt data when read from the L3 cache (or from memory).

[0021] When an exception occurs {e.g., a timer interrupt, a hardware access, etc.) the trust anchor swaps out the key associated with the running VM instance and selects the appropriate key for a new execution context. In one embodiment, the CPU may be configured to signal the trust anchor with an execution context for the next VM instance (or hypervisor) scheduled for execution. For example, the virtualization extensions provided by the Intel® VT-x technology and AMD-V technology. Further, the process for switching keys and initializing an encryption engine included in the trust anchor may require fewer clock cycles than required for the context switch itself.

[0022] The following description is presented to enable one of ordinary skill in the art to make and use the proposed techniques. Descriptions of specific embodiments and applications are provided only as examples and various modifications will be readily apparent to those skilled in the art. The general principles described herein may be applied to other embodiments and applications without departing from the scope of the disclosure. Thus, the present disclosure is not to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features described herein. For purpose of clarity, features relating to technical material that is known in the technical fields related to the proposed ideas have not been described in detail.

1830343J 5 [0023] Figures 1A-1 B are block diagrams illustrating different models for trusted VM execution in Infrastructure as a Service (laaS) cloud environments, according to certain embodiments of the present disclosure. As shown in Figure 1A, a stack model 105 may be used to provide a trusted environment for executing one or more virtual machines 125. In particular, a hardware anchor chip {e.g., a Trusted Platform Module (TPM) 1 10), or equivalent, may be used to perform the following tasks: secure key storage, signature imposition/verification, and secure export of keys, etc. The TPM 1 10 forms the fist link of trust in the stack model 105. Once the TPM 1 10 validates it, the secure BIOS 1 15 may in turn validate that the hypervisor 120 has not been modified. Doing so extends the trust chain from the TPM 1 10 to the secure BIOS 1 15, and subsequently, to the hypervisor 120. Once booted, the hypervisor 120 follows the same procedure to bring up the virtual machine 125. That is, once validated by the secure BIOS 1 15, the hypervisor 120 acts as a trusted actor to validate and boot a virtual machine instance 125. Thereafter, the operating system 130 on each virtual machine 125 do the same to boot applications 135. More simply, TPM 1 10 validates the BIOS 1 15, which validates the hypervisor 120, which in turn validates each VM 125 booted on the hypervisor 120. Thus, the stack model 105 relies on each element in the chain trusting its predecessor.

[0024] In stack model 105, the upper layers rely on the lower layers for trust. If there is any breakage on the chain, the system is not secure. Further, the lower layers on the chain may still have visibility to what upper layers may be executing.

Further still, even when a hypervisor validates and boots a given virtual machine 125, it does not limit the actions performed by trusted OS 130 and trusted applications 135.

Thus, the stack model 105 does not prevent an otherwise validated application 135 (or VM 125) from accessing data belonging to another VM executed by the hypervisor

120, e.g., data stored in a L3 cache.

[0025] Figure 1 B shows an alternative trust model, according to one embodiment presented in this disclosure. As shown, the trust model 150 provides a star model where the trust anchor 165 sits in the middle rather than at the bottom. Doing so allows the trust anchor 165 to directly validate a BIOS 175 and a hypervisor 170 on a virtualization server. But further, doing so also allows the trust anchor 165 to validate each VM image passed to the hypervisor 170 for execution {e.g., VM1 155 and VM2, 160) as well as to validate code for applications 108 to load and execute on the VMs 155,160. That is, unlike the stack model 105, the star model 150 provides a trust

1830343J 6 anchor 165 configured to provides a trust anchor 165. In order to prevent unauthorized access between execution contexts, the star model 150 shifts from integrity protection (in the stack model) to confidentiality for each relevant actor within a virtualization server. Doing so may be used to provide a secure computing environment for VMS 155, 160. For example, unlike the stack model 105, the star model 150 prevents actors with higher privileges {e.g., hypervisor 170) from accessing the data of other actors {e.g., VMs 155,160) as wells as prevents one VM from accessing data belonging to another VM, even when such VMs are executed concurrently in different cores on a multi-core processor and sharing an L3 cache and memory. Figures 2-4, discussed below, provide an example architecture implementing the star model 150 shown in Figure 1 B on a virtualization server with a multi-core CPU.

[0026] Figure 2 is a block diagram illustrating components of a virtualization server 200 configured to provide for the trusted execution of multiple VMs in an laaS environment, according to certain embodiments of the present disclosure. The virtualization server 200 may generally include a collection of hardware computing components connected by one or more interconnects 250. Illustratively, virtualization server 200 includes network interface cards (NICs) 251 , a memory 252, CPU(s) 253, and a storage device 254 {e.g., a locally attached disk drive, solid-state device (SSD), or a connection to remote storage such as a SAN).

[0027] Virtualization server 200 allows multiple virtual machines (VMs) 230i-3 to execute on the server 200 concurrently, sharing the computing hardware present on the virtualization server 200. However, the virtual machines 230-I .3 are not generally aware of the computing hardware on the virtualization server 200 directly. Instead, a hypervisor 240 may be configured to provide virtualized hardware elements for each virtual machine 230i_3. Note, while the virtual hardware allocations appear distinct to the OS and applications 136 running on each virtual machine 230ι_3, they are shared below the virtualization layer. That is, the virtual resources provide an abstraction for the underlying physical resources - and the underlying physical resources are shared among the virtual machines 230i_3.

[0028] Illustratively, the memory 252 includes the hypervisor 240 managing three virtual machine instances (VMs) 230i_3. Each VM 230i_3 provides a virtualized computing platform with a virtual CPU, memory, storage, and networking interfaces.

1830343J 7 An operating system is booted on each VM 230-1 -3 and used to load and execute applications 232-1.3. In one embodiment, the hypervisor 240 may be implemented as a software layer that runs directly on the computing hardware of the virtual ization server 200. In such a case, the hypervisor 240 may be configured to intercept some (or all) operating system calls made by the OS running on a VM 230.

[0029] More generally, the hypervisor 240 runs over the system hardware and allows the virtualization server 200 to host the VMs 230-1.3. In one embodiment, a trust anchor 258 may be used to validate the hypervisor 240. That is, the hypervisor 240 (i.e., the image of the hypervisor 240 on a storage disk) 240 may be signed by an encryption key, and the trust anchor 258 may be configured to obtain the key needed to verify that the hypervisor 240 has not been modified, prior to loading the hypervisor 240. For example, a given laaS provider (or hypervisor vendor) may sign the hypervisor code using a private key of a private/public key pair. In some cases, the public key may also be signed by a certificate authority (and provided as part of a PKI certificate). In turn, the trust anchor 258 may obtain the public key from the PKI certificate and validate the digital signature associated with the hypervisor 240.

[0030] Once validated, the hypervisor 240 may boot and execute VMs 230i_3. Further, each VM 230 may also be a signed and/or encrypted object. For example, an enterprise may transmit a VM image encrypted using a VM encryption session key to the virtualization server 200. In one embodiment, the trust anchor 258 authenticates itself to the enterprise in order to obtain the VM image encryption key associated with a given VM 230 prior to executing that VM 230 on the virtualization server 230. The VM image encryption key may be used to decrypt an encrypted VM image submitted to the virtualization server 200.

[0031] Further, the trust anchor 258 may generate (or obtain) a memory session key for each given VM 230 and use it to encrypt/decrypt information read from and written to the cache 255 while a VM executes on one of the execution cores 256. The trust anchor 258 may be configured to swap memory session keys whenever the execution context on one of the execution cores 256 on CPU 253 is changed, e.g., a context-switch between different VMs 230i_3 or between a VM 230i_3 and the hypervisor 240 itself.

1830343J 8 [0032] The cache 255 provides a high-speed memory accessed by the execution cores 256 on the CPU 253. While memory 252 can be segmented across virtual machines 230, cache 255 is often shared by different VMs 230-1 -3. For example, as shown, CPU 253 may include one or more execution cores 256. When distinct VMs 230-1 -3 are executed on different execution cores 256, the VMs 230^3 may share access to the cache 255. In one embodiment, the trust anchor 258 may use a distinct memory session key for each VM 230 to encrypt/decrypt each read/write to the cache 255. Doing so prevents a VM 230 executing on one execution core 256 from accessing data in the cache 255 belonging to another VM 230 executing on a different execution core 256. Further, doing so also prevents the hypervisor 240 from accessing data in the cache 255 belonging to a given VM 230-I .3, despite the hypervisor 240 having a generally higher privilege level.

[0033] Figure 3 is a block diagram illustrating an example of a processor architecture configured to provide for the trusted virtual machine execution in an laaS environment, according to certain embodiments of the present disclosure. As shown, a CPU 300 includes two distinct processing cores 305, 310. And each CPU core 305 includes a layer 1 (L1 ) instruction and data cache and an L2 cache. CPU 300 also includes a trust anchor 315, and processing cores 305, 310 each have an associated key generation component 320i_2 and an encrypt/decrypt block 325i_2. Illustratively, cache lines connecting processing cores 305, 310 to a shared L3 cache 330 each include one of the encrypt/decrypt blocks 325-I .2. The L3 cache 330 is connected to a memory controller 335, which in turn is connected to a memory 340.

[0034] Although CPU 300 illustrates an example embodiment with two processing cores 305, 310, one of ordinary skill in the art will recognize that the embodiments described herein can readily be adapted for a CPU having more processing cores. In such a case, each additional CPU core would include a connection to the trust anchor 315 as well as a key generation component 320 an encrypt/decrypt block connected the shared L3 cache 330.

[0035] In one embodiment, the trust anchor 315 may be configured to manage multi-tenancy execution context (i.e., the execution of multiple virtual machine images) on the processing cores 305, 310 as well as manage communications with the external world. For example, the trust anchor 315 may be configured to provide a secure boot process for loading and booting a VM instance, to provide for secure

1830343J 9 context switching between VM instances, to swap memory session keys based on execution context, and to provide for secure key storage during VM execution. Further, the trust anchor 315 embedded with the CPU 300 may be provisioned, at manufacturing time, with a public/private key pair and a certificate issued by a certificate authority (CA) that binds the trust anchor identity to the public key. The CPU model and manufacturer name may be included as attributes of this certificate. In one embodiment, the trust anchor 315 may use the public/private key pair to prove that a virtualization server has a multi-core CPU configured with a trust anchor and, accordingly, can boot and execute a VM image in an otherwise untrusted cloud environment.

[0036] Additionally, the trust anchor 315 may receive an indication from a processing core 305, 310 whenever a context switch occurs. Such an indication may provide an identifier for an incoming execution context. As noted above, certain multi- core processors provide an extended instruction set that allow for an execution context to be exported from a processing core on a CPU. For example, the virtualization extensions provided by the Intel® VT-x technology and the VMX instruction set provide instructions used to signal changes in execution context on a processing core {e.g., the VMEXIT and VMRESUME instructions).

[0037] In response to a signal from one of the cores 305, 310 indicating a change in execution context, the trust anchor 315 may configure the encrypt/decrypt block

325-I -2 and key generation component 310-1-2 associated with that processing core 305, 310 for the new execution context. When a context switch occurs, the trust anchor swaps out a key associated with the then current execution context on a processing core 305, 310 and replaces it with a key associated with the incoming execution context. As is known, a context switch on a multi-core processor between VM instances (or between a VM instance and the hypervisor) requires registers {e.g., status, IR, general purpose, counters) on the CPU core to be restored (or initialized) for the incoming execution context. In one embodiment, the trust anchor 315 may be configured to swap the appropriate encryption key and initialize the encrypt/decrypt block 325-I .2 using fewer clock cycles than required to perform the context switch itself (currently, a process that typically requires -20-30 clock cycles). Doing so allows the trust anchor to provide a secure execution context for the incoming VM instance (or hypervisor) without introducing any collateral latency.

1830343J 10 [0038] Once initialized, the encrypt/decrypt block 325 encrypts/decrypts data as it is moved over cache lines between the processing core 305, 310 and the L3 cache 255. In one embodiment, the encrypt/decrypt blocks 325i_2 provide a hardware based implementation of a stream cipher. As is known, stream cipher is a symmetric key cipher where plaintext is combined with a keystream to encrypt small units of data {e.g., a single bit or byte at a time). In a particular embodiment, the encrypt/decrypt blocks may be configured to use AES-CTR (Advanced Encrypted Standard-Counter mode) as the stream cipher. Using a stream cipher allows data to be encrypted/decrypted as it moves between a processing core 305, 310 and the L3 cache 330 without requiring additional clock cycles. Instead, once initialized, the encrypt/block block 325i_2 may simply XOR each bit being moved to/from the particular core 305, 310 using the keystream. Thus, the functionality provided by the trust anchor is located adjacent to each processing core 305, 310 and secures any element shared by VMs running on different cores (in this case L3 cache 330 and memory 340). In one embodiment, the key generation components 320i_2 may be configured to use a VM key, a memory address and a nonce to generate a keystream used by the encryption/decryption blocks 325i_2.

[0039] The memory controller 335 may be configured to move data between memory 340 and the L3 cache. For example, the memory controller 335 may select to flush data from the L3 cache 330 to the memory 340. In one embodiment, when doing so, the memory controller 335 also writes the nonce value used to encrypt the data when it was written to the L3 cache 330 to memory 340. Similarly, the memory controller 335 may retrieve the appropriate nonce value when reading encrypted data from the memory 340.

[0040] Figure 4 further illustrates aspects of the processor architecture first shown in Figure 3, according to certain embodiments of the present disclosure. In particular, Figure 4 shows the processor architecture of Figure 3 used to perform a secure write operation 401 to an L3 cache 433 and used to perform a secure read operation 402 from the L3 cache 433. [0041] As shown, the write operation 401 encrypts plaintext data 405i as it is moved from an L2 cache local to a processing core to an L3 cache 433 shared by multiple processing cores. Illustratively, an encrypt/decrypt block 410i includes a key

1830343J 1 1 415-1 , an AES engine 420 , a memory address 425 and a nonce 430 . The key 415 is provided to the encrypt/decrypt block 41 O by the trust anchor. As noted above, the key 415i corresponds to a VM (or hypervisor) being executed on one of the processing cores of a multi-core CPU {e.g., the processor architecture shown in Figure 3). The memory address 425 corresponds to a memory address of the plaintext 405 being written from an L2 cache to the shared L3 cache 433. The nonce 430i provides a one-time value used for a given write operation. Typically, the nonce 430 may be a counter value incremented for each write (and read) to the shared L3 cache 433. Alternatively, however, the nonce 430 may be generated from a random or pseudo-random noise source.

[0042] In one embodiment, the AES engine 420 provides a hardware implementation of the AES-CTR stream cipher and is configured to generate a keystream from the key material, i.e., from the key 415-1 , the memory address 425 , and the nonce 430 . The resulting keystream is XORed with the plaintext 405 as it is pushed from the L2 cache on a processing core to the shared L3 cache 433, resulting in ciphertext 435 . In addition to writing the resulting encrypted data to the L3 cache 433, the nonce value in the encrypt/decrypt block 410 is also written to the L3 cache 433 (shown in Figure 4 as nonce 440 ). In one embodiment, the encrypt/decrypt block 410 may include a separate cache line to the L3 cache 433 allowing the nonce 430 to be stored in the L3 cache 433 (as nonce 440 ) in parallel with the plaintext 405 being enciphered using the keystream and stored in the L3 cache 433 as ciphertext 435 . Alternatively, the nonce 340 may written to the cache 433 using the same cache line used to write the ciphertext 435 .

[0043] The read operation 402 is similar to the write operation 401 , but in the reverse direction. To read enciphered data 4352 from the L3 cache 433, the memory address 4252 and the nonce 4402 are read into the encrypt/decrypt block 4102. Again, the key 4152 is the session key associated with a VM (or hypervisor) being executed on a processing core of a multi-core processor - and remains unchanged between context switches on the processing core. The memory address 4252 corresponds to the memory address of the data being pulled into the L2 cache and

1830343J 12 the nonce 4302 is the nonce value used to encipher the data when it was written to the L3 cache 433. That is, the same key material used by the write operation 401 to generate a keystream used to encrypt the plaintext 405i is used to re-generate the same keystream to decrypt the encrypted data 4352 from the L3 cache 433. The resulting keystream is then XORed with the encrypted data 4352 from the L3 cache 433 and written as plaintext 4052 in the L2 cache.

[0044] Figure 5 illustrates a method 500 for securely booting a VM instance, according to certain embodiments of the present disclosure. As shown, the method 500 begins at step 505 where a hypervisor receives a boot image of a virtual machine to boot and execute on a virtualization server. For example, in one embodiment, a user may transmit a VM boot image to a cloud provider along with a set of requirements for the virtualized hardware system on which to execute the VM instance {e.g., a processor speed, memory requirements, etc.). The VM boot image may be encrypted using a VM image encryption key generated by the user. In one embodiment, each 4kb chunk is encrypted independently so that chunks may be decrypted and paged in and out of memory on the virtualization server without decrypting the whole image

[0045] At step 510, the hypervisor attempts to boot the VM boot image. At step 515, the hypervisor determines whether the VM boot image is encrypted. If so, the hypervisor triggers a negotiation between the trust anchor and the user that submitted the VM boot image for execution (or an associated key service) to obtain the VM image encryption key used to encrypt the VM boot image. Once obtained, the trust anchor configures a disk control block with that key (step 520). The VM image encryption key is used to decrypt memory pages of the encrypted VM boot image as they are fetched from storage and stored in memory (or encrypt pages flushed from memory to storage). The trust anchor stores the VM image encryption key associated with the VM image in a secure key store. At step 525, the trust anchor also generates a memory session key used to configure an encrypt/decrypt block in cache lines between an L3 cache and memory and between the L3 cache and an L2 cache on each processing core. Once the encrypt/decrypt blocks on the CPU are configured, the VM image may be fetched from storage and booted. The hypervisor may them boot the VM image and schedule it for execution (step 530).

1830343J 13 [0046] At step 535, the hypervisor executes the VM image on one of the processing cores. As portions of the encrypted VM image are fetched from storage, the disk control block decrypts them using the VM image encryption key (step 540). The memory session key is then used to re-encrypt such elements as they are written to an L3 cache. When data for a given memory address is needed, the memory session key in the cache line between the L2 cache of a processing core and the shared L3 caches is used to decrypt data read from (and written to) the L3 cache. Thus, the data in the L3 cache (shared among multiple processing cores) remains encrypted until fetched into an L2 cache (local to a processing core).

[0047] If the VM boot image is not encrypted (step 515), then the trust anchor may still generate a memory session key in order to encrypt/decrypt all write/reads to the L3 cache made by the VM while executing on a processing core (step 525). In either case, the hypervisor may begin scheduling the VM for execution on one of the processing cores of a multi-processor CPU (step 530). Additionally, when the hypervisor initiates a context switch, the trust anchor configures the encrypt/decrypt blocks on the CPU with the appropriate VM image encryption key and memory session key for an incoming VM.

[0048] Figure 6 illustrates a method 600 for performing a context switch on a processing core executing multiple VM instances, according to certain embodiments of the present disclosure. As shown, the method 600 begins at step 605, where a next VM instance (or hypervisor) is selected for execution on a processing core of a CPU. For example, an exception may occur while one VM instance (or hypervisor) is being executed on the processing core {e.g., a timer interrupt, a cache miss, hardware access, etc.). When this occurs, a context switch may be initiated to park the VM instance (or hypervisor) then running on the processing core and schedule a next VM instance (or hypervisor) for execution on that core.

[0049] In one embodiment, the processing core may signal the trust anchor with an indication of an execution context for the VM instance (or hypervisor) being restored to the processing cores (i.e., of an incoming execution context). For example, as noted above, the processing cores may execute the VMEXIT and VMRESUME instructions to signal the trust anchor of the exiting and incoming execution contexts.

1830343J 14 [0050] At step 610, the trust anchor stores the session key associated with the current execution context while the VM instance (or hypervisor) is being parked and the incoming execution context is initialized. At step 615, the execution context of the selected VM instance (or hypervisor) is restored. That is, the CPU restores any register values and associated system state needed to resume execution of a VM instance (or hypervisor) associated with the incoming execution context.

[0051] At step 620, while the current execution context is parked and the incoming execution context is restored, the trust anchor may retrieve a session key for the VM instance (or hypervisor) associated with the execution context being restored. And the trust anchor may use that session key to initialize an encrypt/decrypt block associated with the processing core performing the context switch. Once the session key is restored, and the encrypt/decrypt block is initialized, the VM instance (or hypervisor) may begin executing on the processing core. As noted above, each read/write operation moving data from/to the L3 cache may then be encrypted using the session key and key material {e.g., a memory address of any read/write operation to the L3 cache and a nonce value). Thus, memory writes performed by the incoming execution context are encrypted with a key unique to that context. Configuring the trust anchor to swap such keys based on the execution context ensures the compartmentalization of the memory of each individual execution element (i.e., of the given VM instance or the hypervisor) executed on the processing core(s) of a CPU. Thus, even if a given VM instance attempts to access an L3 cache line of another, it does not have access to the key needed to decrypt that cache line. Further, the same result would occur if the hypervisor attempted to read data stored in the L3 cache by a given VM instance, despite the hypervisor generally having a higher privilege level. In either case, as the hypervisor (and VM instances) do not have access to the session key of other VM instances, the attempted access would result in garbage data.

[0052] In sum, embodiments described herein provide techniques for securely executing multiple virtual machine instances on a multi-core CPU. A trust anchor may be configured to manage keys used to encrypt/decrypt instructions and data when executed on a core of the multi-core CPU as well as perform the encryption/description operations. When a context switch occurs, the trust anchor may restore a key (along with other data) to encrypt/decrypt memory read to and written from a shared L3 cache. In one embodiment, the CPU may signal the trust

1830343J 15 anchor with an execution context for the next VM instance (or hypervisor) scheduled for execution.

[0053] Advantageously, configuring a multi-core CPU with a trust anchor as described herein enables the trusted execution of VM instances, even in non-trusted cloud environments and on non-trusted hypervisors. That is, embodiments described herein provide a trusted execution and memory compartmentalization framework for execution of VM in third-party cloud providers independently from the provider itself being trusted or not. In one embodiment, a CPU architecture may be configured to replace a traditional trusted chain approach of TPM based architectures with a star-of- trust architecture where the CPU is at the center of the star.

[0054] While the foregoing is directed to embodiments of the present disclosure, other and further embodiments of the disclosure may be devised without departing from the basic scope thereof, and the scope thereof is determined by the claims that follow.

1830343J 16

Claims

WE CLAIM:
1 . A method for performing a context switch on a processing core of a CPU, comprising:
receiving an indication of an incoming execution context for the processing core;
storing a first session key, wherein the first session key is associated with a current execution context of the processing core;
retrieving a second session key, wherein the second session key is associated with the incoming execution context; and
initializing an encrypt/decrypt block on the CPU using the second session key, wherein the encrypt/decrypt block is disposed between the processing core and a cache accessed by the processing core.
2. The method of claim 1 , further comprising, initiating the incoming execution context on the processing core, wherein the encrypt/decrypt block either encrypts or decrypts data read from the cache using the second session key.
3. The method of claim 1 , wherein the CPU includes multiple processing cores and wherein each processing core shares access to the cache.
4. The method of claim 1 , wherein the first session key and the second session key are stored by a hardware trust anchor coupled to the CPU.
5. The method of claim 1 , wherein the incoming execution context references a first virtual machine (VM) instance to be executed on the processing core following the context switch.
6. The method of claim 1 , wherein the incoming execution context references a hypervisor to be executed on the processing core following the context switch.
7. The method of claim 1 , wherein the encrypt/decrypt block provides a hardware implementation of a stream cipher, and wherein a key stream is generated for one of a read from the cache using the second session key, a nonce value, and a memory address associated with the read operation and a write operation to the cache using the second session key, a nonce value, and a memory address associated with the write operation.
1830343J 17
8. A computing system, comprising,
a memory; and
a CPU having:
a processing core,
a cache, and
a trust anchor, wherein the trust anchor is configured to:
receive an indication of an incoming execution context for the processing core;
store a first session key, wherein the first session key is associated with a current execution context of the processing core;
retrieve a second session key, wherein the second session key is associated with the incoming execution context; and
initialize an encrypt/decrypt block on the CPU using the second session key, wherein the encrypt/decrypt block is disposed between the processing core and a cache accessed by the processing core.
9. The computing system of claim 8, wherein the processing core is configured to, following the initialization of the encrypt/decrypt block, initiate the incoming execution context on the processing core, wherein the encrypt/decrypt block decrypts data read from the cache using the second session key, and wherein the encrypt/decrypt block encrypts data written to the cache using the second session key.
10. The computing system of claim 8, wherein the CPU includes a plurality of processing cores and wherein each processing core shares access to the cache.
1 1 . The computing system of claim 8, wherein the first session key and the second session key are stored by the hardware trust anchor.
12. The computing system of claim 8, wherein the incoming execution context references a first virtual machine (VM) instance to be executed on the processing core following a context switch.
13. The computing system of claim 8, wherein the incoming execution context references a hypervisor to be executed on the processing core following a context switch.
1830343J 18
14. The computing system of claim 8, wherein the encrypt/decrypt block implements a stream cipher, and wherein a key stream is generated for one of a read from the cache using the second session key, a nonce value, and a memory address associated with the read operation and a write operation to the cache using the second session key, a nonce value, and a memory address associated with the write operation.
1830343J 19
PCT/US2011/037799 2010-05-27 2011-05-24 Method and apparatus for trusted execution in infrastructure as a service cloud environments WO2011149983A3 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US12/789,189 2010-05-27
US12789189 US8812871B2 (en) 2010-05-27 2010-05-27 Method and apparatus for trusted execution in infrastructure as a service cloud environments

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN 201180036949 CN103038746B (en) 2010-05-27 2011-05-24 Methods for Infrastructure as a Service cloud environment and credible implementation of the device
EP20110787280 EP2577449A4 (en) 2010-05-27 2011-05-24 Method and apparatus for trusted execution in infrastructure as a service cloud environments

Publications (2)

Publication Number Publication Date
WO2011149983A2 true true WO2011149983A2 (en) 2011-12-01
WO2011149983A3 true WO2011149983A3 (en) 2012-03-29

Family

ID=45004732

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2011/037799 WO2011149983A3 (en) 2010-05-27 2011-05-24 Method and apparatus for trusted execution in infrastructure as a service cloud environments

Country Status (4)

Country Link
US (1) US8812871B2 (en)
EP (1) EP2577449A4 (en)
CN (1) CN103038746B (en)
WO (1) WO2011149983A3 (en)

Families Citing this family (27)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB201312923D0 (en) * 2010-12-21 2013-09-04 Ibm Virtual machine validation
US8495356B2 (en) * 2010-12-31 2013-07-23 International Business Machines Corporation System for securing virtual machine disks on a remote shared storage subsystem
GB2501433B (en) * 2011-01-19 2014-06-04 Ibm Ensuring the integrity of software
US8799997B2 (en) * 2011-04-18 2014-08-05 Bank Of America Corporation Secure network cloud architecture
US8813192B2 (en) * 2011-05-04 2014-08-19 Novell, Inc. Techniques for establishing a trusted cloud service
US8943564B2 (en) * 2011-07-21 2015-01-27 International Business Machines Corporation Virtual computer and service
US20130166697A1 (en) * 2011-12-22 2013-06-27 Gregory P. Manning Multiconfiguration device cloud entity protocol
US8838968B2 (en) 2012-05-14 2014-09-16 Ca, Inc. System and method for virtual machine data protection in a public cloud
US8817707B2 (en) * 2012-07-20 2014-08-26 Intel Corporation Mechanisms for roaming between 3GPP operators and WLAN service providers
US9262188B1 (en) * 2012-09-19 2016-02-16 Emc Corporation Virtual asset management in data center
CN104756127B (en) 2012-10-12 2018-03-27 皇家飞利浦有限公司 Secure data processing by the virtual machine
US9069766B2 (en) * 2012-11-02 2015-06-30 Microsoft Technology Licensing, Llc Content-based isolation for computing device security
US9298502B2 (en) * 2013-01-31 2016-03-29 Empire Technology Development Llc Pausing virtual machines using API signaling
WO2014142858A1 (en) * 2013-03-14 2014-09-18 Intel Corporation Trusted data processing in the public cloud
US9292331B2 (en) * 2013-03-15 2016-03-22 Bracket Computing, Inc. Expansion of services for a virtual data center guest
US9171163B2 (en) 2013-03-15 2015-10-27 Intel Corporation Mutually assured data sharing between distrusting parties in a network environment
US9652331B2 (en) * 2013-10-24 2017-05-16 Sas Institute Inc. Techniques for creating a bootable image in a cloud-based computing environment
US20150244716A1 (en) * 2014-02-24 2015-08-27 Amazon Technologies, Inc. Securing client-specified credentials at cryptograpically attested resources
US9503433B2 (en) * 2014-03-27 2016-11-22 Intel Corporation Method and apparatus for cloud-assisted cryptography
EP2978159A1 (en) * 2014-07-21 2016-01-27 Nxp B.V. Nonce generation for encryption and decryption
US9553721B2 (en) * 2015-01-30 2017-01-24 Qualcomm Incorporated Secure execution environment communication
US9954681B2 (en) * 2015-06-10 2018-04-24 Nxp Usa, Inc. Systems and methods for data encryption
US9781163B2 (en) * 2015-06-23 2017-10-03 EMC IP Holding Company LLC Trust overlays for disaggregated infrastructures
US9667606B2 (en) 2015-07-01 2017-05-30 Cyphermatrix, Inc. Systems, methods and computer readable medium to implement secured computational infrastructure for cloud and data center environments
US9894061B2 (en) 2015-10-16 2018-02-13 International Business Machines Corporation Method for booting and dumping a confidential image on a trusted computer system
US20170208149A1 (en) * 2016-01-20 2017-07-20 International Business Machines Corporation Operating local caches for a shared storage device
CN105790932A (en) * 2016-02-25 2016-07-20 四川长虹电器股份有限公司 Encryption method through using machine codes as bases

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1126356A2 (en) 2000-02-14 2001-08-22 Kabushiki Kaisha Toshiba Tamper resistant microprocessor

Family Cites Families (21)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6101255A (en) * 1997-04-30 2000-08-08 Motorola, Inc. Programmable cryptographic processing system and method
US6895506B1 (en) * 2000-05-16 2005-05-17 Loay Abu-Husein Secure storage and execution of processor control programs by encryption and a program loader/decryption mechanism
JP4098478B2 (en) * 2001-01-31 2008-06-11 株式会社東芝 Microprocessor
JP2002353960A (en) * 2001-05-30 2002-12-06 Fujitsu Ltd Code performing device and code distributing method
JP2003051819A (en) * 2001-08-08 2003-02-21 Toshiba Corp Microprocessor
US7272832B2 (en) * 2001-10-25 2007-09-18 Hewlett-Packard Development Company, L.P. Method of protecting user process data in a secure platform inaccessible to the operating system and other tasks on top of the secure platform
JP3880933B2 (en) * 2003-01-21 2007-02-14 株式会社東芝 Data access control method according to the tamper resistant microprocessor and a cache memory mounted processor
JP2004287590A (en) * 2003-03-19 2004-10-14 Matsushita Electric Ind Co Ltd Debugging system, microprocessor, and debugger
US7222062B2 (en) 2003-12-23 2007-05-22 Intel Corporation Method and system to support a trusted set of operational environments using emulated trusted hardware
DE102004014435A1 (en) * 2004-03-24 2005-11-17 Siemens Ag Assembly having an integrated circuit
JP4447977B2 (en) * 2004-06-30 2010-04-07 富士通マイクロエレクトロニクス株式会社 Secure processor, and a program for the secure processor.
US7613921B2 (en) 2005-05-13 2009-11-03 Intel Corporation Method and apparatus for remotely provisioning software-based security coprocessors
WO2007063437A3 (en) * 2005-11-29 2009-12-10 Nxp B.V. Storage media
US8001374B2 (en) * 2005-12-16 2011-08-16 Lsi Corporation Memory encryption for digital video
CN101004776B (en) * 2006-01-09 2010-06-16 太阳微系统有限公司 Method and apparatus for protection domain based security
EP1870814B1 (en) * 2006-06-19 2014-08-13 Texas Instruments France Method and apparatus for secure demand paging for processor devices
WO2010019916A1 (en) * 2008-08-14 2010-02-18 The Trustees Of Princeton University Hardware trust anchors in sp-enabled processors
CN101350044B (en) * 2008-09-02 2010-07-14 中国科学院软件研究所 Method for constructing virtual environment trust
US7870044B2 (en) 2008-10-02 2011-01-11 Verizon Patent And Licensing Inc. Methods, systems and computer program products for a cloud computing spot market platform
US8738932B2 (en) * 2009-01-16 2014-05-27 Teleputers, Llc System and method for processor-based security
US8719593B2 (en) * 2009-05-20 2014-05-06 Harris Corporation Secure processing device with keystream cache and related methods

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1126356A2 (en) 2000-02-14 2001-08-22 Kabushiki Kaisha Toshiba Tamper resistant microprocessor

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP2577449A4

Also Published As

Publication number Publication date Type
CN103038746B (en) 2017-03-29 grant
WO2011149983A3 (en) 2012-03-29 application
US8812871B2 (en) 2014-08-19 grant
CN103038746A (en) 2013-04-10 application
EP2577449A4 (en) 2017-07-05 application
EP2577449A2 (en) 2013-04-10 application
US20110296201A1 (en) 2011-12-01 application

Similar Documents

Publication Publication Date Title
McCune et al. Flicker: An execution infrastructure for TCB minimization
Arnautov et al. SCONE: Secure Linux Containers with Intel SGX.
Keller et al. NoHype: virtualized cloud infrastructure without the virtualization
Lie et al. Implementing an untrusted operating system on trusted hardware
US8108641B2 (en) Methods, apparatus, and systems for secure demand paging and other paging operations for processor devices
US20120331550A1 (en) Trusted language runtime on a mobile platform
Azab et al. Sice: a hardware-level strongly isolated computing environment for x86 multi-core platforms
US8156298B1 (en) Virtualization-based security apparatuses, methods, and systems
US8656482B1 (en) Secure communication using a trusted virtual machine
US20130097356A1 (en) System and method for kernel rootkit protection in a hypervisor environment
US20090164994A1 (en) Virtual computing management systems and methods
US20120102307A1 (en) Method and apparatus including architecture for protecting sensitive code and data
US7613921B2 (en) Method and apparatus for remotely provisioning software-based security coprocessors
US20050021944A1 (en) Security architecture for system on chip
US20120159184A1 (en) Technique for Supporting Multiple Secure Enclaves
Koeberl et al. TrustLite: A security architecture for tiny embedded devices
Costan et al. Intel SGX Explained.
US7073059B2 (en) Secure machine platform that interfaces to operating systems and customized control programs
US7571312B2 (en) Methods and apparatus for generating endorsement credentials for software-based security coprocessors
US20070094719A1 (en) Method and apparatus for migrating virtual trusted platform modules
US20060256106A1 (en) Method and apparatus for migrating software-based security coprocessors
US20060256105A1 (en) Method and apparatus for providing software-based security coprocessors
US20040177243A1 (en) Customized execution environment
Champagne et al. Scalable architectural support for trusted software
US20090125974A1 (en) Method and system for enforcing trusted computing policies in a hypervisor security module architecture

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 11787280

Country of ref document: EP

Kind code of ref document: A2

NENP Non-entry into the national phase in:

Ref country code: DE